Voltage Stabilizer Notes
First published 05/02  R. Kwas  Revisions ongoing [ Comments added. ] 

Update - Sep2018 

---------------------------

General:   In order to assure accuracy of the instruments and independence from large vehicle voltage system (power buss) variations, and also to prevent instantaneous fluctuations to be visible on gauges, a strategy of using slow response gauges along with a Voltage Stabilizer (also referred to as VS or V-Stab here, and not to be confused with the Voltage Regulator which controls the Charging System!) was employed on late 1800 (injected) and 140 Volvo vehicles.  The component for 1800s was PN 686396 ("10V") and was supplied by manufacturer Smiths, along with the instruments of the 1800s.  Instruments of the 140s (164s) were VDO supplied by a V-Stab with PN 684623 ("5.1V").  In operating principle, they are quite similar, however, they are not interchangeable, and this article will focus on the V-Stabs as used in the 1800s.    

In the late sixties, some of the gauges were implemented using a thermo-mechanical module which contained a cycling Bi-metalic element.  The original Smiths design was OK [maybe better than just OK!  See below **] for its day...bimetalic element mechanisms are predictable, repeatable and reliable, but combining these with a heating element, with its connections of dissimilar metals, starts to reduce that reliability...add a few decades and the design can develop problems and fail. 


Excerpt from Volvo Service Manual for 1971 1800E

 

"Constant", vs. Average, vs. Continuous.  According to the factory manual, the output switches from 12V to 0V, with the long-term average value being 10V (this is incorrectly called "constant" in the manual text, possibly due to quirks in translation from Swedish).  The important point to note is that this long-term average voltage value is a function of % Duty Cycle, defined as the ratio of ON voltage and OFF voltage time.

Solving for Duty Cycle Percentage = 10/12 X 100 = 83.33%

In other words, the output is 12V, 83.33% of the time, and 0V, 16.67% of the time, and quite independent of the actual switching frequency (notice there in no frequency factor in the duty-cycle calculation(!), but just for reference, the typical actual ON-time is approx 0.8sec.,  and OFF-time is approx. 0.2, making the operating frequency about 1Hz, but again, totally non-critical as long as the duty cycle is 83% (concerns that accuracy suffers with varying ambient temps are highly overstated, and just plain wrong!  The frequency is higher at hot ambient temps, but Duty-Cycle remains correct with this wonderfully tolerant design!  See below!).  As long as this ratio is correct, the long-term effective voltage is 10 +/- .2V, and instruments will operate just fine and retain their accuracy! 


VS output voltage plotted against time axis. Note the initial ON-Time is much longer than that of the normal cycle once settled.  This can easily be explained by the fact that first time the bi-metal has to open the contact, it had to be heated from ambient, later, the element only needs to cycle between opening and closing temperatures.  Math specialists may also note that the "Area Under the Curve", or long-term average as the author prefers to call it, levels out at the desired 10V!  

 

V-Stab failure modes:  The typical failure mode is that the Heating Element is open or otherwise no longer connected and causing the internal contact to remain permanently closed, and this includes an open chassis connection!  This applies a continuous 12V to the output terminal and in-turn to the three instruments on that node.  Instruments will read high values as a result, but are not damaged as a rule.  That instruments are damaged by applying continuous 12V, when their normal operating condition is that they have 12V applied for more than 3/4 of the time, is more questionable information out there...I just fail to believe an instrument which spends it entire operational life seeing 12V 83% of the time will somehow be damaged when it sees this same voltage 100% of the time or continuously...I just don't believe it!!  I am quite certain that Smiths engineers considered this failure Mode during their analysis, and made the system (i.e. Instruments), quite tolerant of it!  In fact I can just about guarantee it! 

----------------

VS Wiring:  As can be seen in the wiring diagram following, terminal B of VS is the IGN Power input from Fuse 4.  Terminal I is the output and this terminal powers items 27 (Fuel Gauge), 29 (Oil Temp), 66 (Coolant Temp).  It is quite important for the VS to have a good electrical connection at its mechanical mount, as in addition to mounting, this also supplies a current path to the chassis for the internal Heating Element, critical to its correct operation!  A loose, poor electrical connection just at the mount would result in non-function of the V-Stab! 

Failure of VS will affect only the meters powered by V-Stab!  I have heard of meters being damaged by a continuous 12V from a failed VS, but I have not confirmed this as a typical or valid failure senario...since the instruments are being powered by 12 and 0V continually alternating, I rather doubt that applying 12V continuously could result in damage...whacky or generous readings sure, but not damage...any reader having confirmed info to the contrary is welcome to contact the author and try to convince me of this! 


Excerpt from factory 1971 1800E Wiring Diagram.  Voltage Stabilizer and related components and wiring highlighted. 

 


(Corrected) Voltage Stabilizer with internal details. 
Note Chassis Tab (C) is bent around board and makes contact with case there...so
the Mount (M) also serves as the electrical connection crucial for correct function.

The most common failure mode for the VS is failure to cycle and stuck closed (steady voltage output terminal equal to input) when the heating element fails to heat.  This can be caused by the element having burned open, corrosion at the connection end(s) of heating wire preventing current flow, or corrosion at the chassis tab which is crimped over the board edge and makes contact when case is crimped over that [see good example of that below!], also preventing current flow, or simply a loose securing thumbscrew which will also inhibit the chassis return current of the Heating Element! 

--------------------------------------------------

If you confirm the V-Stab itself has failed, and feel adventurous enough to gently uncrimp the housing, to do an internal investigation, you may be able to determine the exact cause of failure, and correct this...otherwise a replacement is due.  Preferring a possibly more reliable, modern all electronic solution, I would never reinstall a thermo-mechanical unit.  This is an ideal use for a highly reliable, fully electronic regulator putting out a continuous (truly constant!), not just average 10VDC.  The instruments draw only a small amount of current so represent a minimal load...a 1Amp regulator would essentially be only loafing along totally understressed.  I suggest using a completely electronic 1A regulator, such as an LM7810 (fixed) or LM317 (adjustable, set permanently to 10V), or even the fixed V low drop-out LM2940T-10.0 is referenced in the BritishV8 link below . It takes a bit of electronic micro-construction, but I have even fit these into the original metal case, so when reinstalled, the upgrade is not even obvious without a close inspection, or electrical test.  Sorry, no pix available of this... externally, it wouldn't even look any different from the unmodified unit anyway! Read On!

---------------------------------------------------

Update - Sep2018:  Having just received a failed unit from Graham W. of the VOC Forum (See Thread - Voltage Stabilizer:  https://www.volvoforums.org.uk/showthread.php?t=284381  ), its time for a more in-depth study to reveal the darkest secrets of the original design, and maybe also show a SW-EM updated version.


V Stab as received from Gordon in Britain, nicely protected in a soft envelope...knowing that this component lives fairly well protected,
behind the dashboard, one has to ask if this car spent some time not just near, but in, the English Channel!

 


Initial Resistance tests

Chassis to Term B = 118Ohms (Resistance of the Heating Element.)
Chassis to Term I = 118Ohms
Term I to Term B = 0.7Ohms (Indicative of a closed output contact.)

        (Quick) calculations (based on Ohms Law):  P = V2 / R =  12 X 12 / 120 = 1.2Watts is the power dissipated by the Heating Element when under power.  Naturally this is decreased by the 83% factor. 

 

Heating Element tab view close-up.  Since the 118Ohms (Resistance of the Heating Element) is measurable, I expect this unit would function as intended (bench tests verified this, see below!), so how come Gordon experienced symptoms of "not reading correctly"?  See also explanations below for answers to this question! 


Highlighted in Yellow is the (C) chassis tab by which internal (H) Heating Element connects to chassis (simply under the crimped cover).  It, along with the case looks pretty poor and one would have to question if a good electrical connection exists between the two...but continuity and functional tests confirm:  It does!  This unit, although looking terrible, worked perfectly in tests!

If this connection were open, the HE would not pass current to make heat, so there would be no action of the Bimetal, and contact will remain permanently closed.  That is a good probable explanation of failure even before the component is opened! [Confirmed!]  

-------------

Bench Tests, explaining the Failure Mode, Disassembly, Reassembly with Solid State Electronic Technology (no moving parts!):  

Gordon's old V-Stab was set up for operational tests with a DC power supply set to 13.2V (a reasonable power bus voltage for a vehicle with running engine/charging system).  Power was applied across the input B terminal and mounting tab, while voltage at output I terminal was checked.  Result:  The unit under test functioned perfectly!...within a few seconds of first applying power, voltage at the output terminal started to dutifully cycle  between 12 and 0V at a frequency of about 1Hz.  This brings up the question of why Gordon was experiencing instrument failure symptoms of "not reading correctly" [He later clarified that Temp and Fuel Gauges read high.]

Outside the possible effects of an intermittent operation this can be explained simply, and with probably one of the most common issues on vehicles:  An open chassis connection! ...and since this component's chassis connection is made by the mounting hardware, I would say that it was simply loose. 

Recreation and Explanation of Failure Symptoms:  Indeed, while connected and operating perfectly, the Mounting tab chassis connection was opened.  This resulted in stoppage of the thermal cycling...the output contact closed, and remained closed putting full voltage on the output terminal.  The result of this open chassis connection when installed in Gordon's vehicle and with Instruments being powered, would have similarly been powering the Instruments with a full 12V, and generous readings, certainly consistent with his reports.    

Gordon's replacement action naturally meant loosening the (thumbscrew) nut which secures the V-Stab, (and more importantly completes its current path to chassis), installing the new V-Stab and retightening the nut, so installing the replacement certainly fixed the problem...but so would have putting a dab of ACZP under the nut and retightening...just saying...!  [ Gordon may not be happy that he will have to revisit the back of his 1800ES Dashboard, but I have sent him some ACZP for the mounting stud of the replacement V-Stab.  Once applied, I expect the proper operation of the V-Stab will outlive all of us!]    

Component dissection and reengineering:  Although functioning well, a V Stab in this condition should never be installed into a vehicle, because that contact tab could very well fail any day, so it will be subjected to further performance tests, including during temperature and mounting attitude variations (two parameters alleged to affect operation), and finally a conversion to full electronic.

Special Functional Tests: 

PLACEHOLDER FOR Thermal Tests:

PLACEHOLDER FOR Mounting Attitude Tests:

 

PLACEHOLDER FOR Electronic Conversion:  [I have seen reports of failures of these, likely due to lack of heat-sinking which allows the electronic regulator to overheat causing failure...to include adequate heat-sinking is very important to assure a long life of the electronic component!  Further:  In the original design, heat is generated by the HE, but the voltage from B to I is either 0 or 12, meaning the working contact is operating in a "digital mode" full ON or full OFF, and while OFF, contributes little heat to the assembly.  In the solid-state conversion putting out a steady 10V, the electronics, namely the pass transistor of the regulator, is operating in a linear mode, where it is dissipating power constantly, and this can be calculated as a function of Voltage-drop (nominal 2V) multiplied by load current (exact value ?, and this can vary from vehicle to vehicle based on number of instruments downstream and being supplied by V-Stab)...point is that this heat must be shed from the electronic regulator or it will get too hot and its operating life will be shortened...in that case, a bimetallic element might be more reliable...I never expected to say that! ] 

---------------------------------------------------

Links to discussions on the Subject:

http://www.brickboard.com/RWD/volvo/1321934/1800/fuel_gauge_works_oil_coolant_temp_gauges_dont_voltage_stabilizer_1800e.html

http://www.britishv8.org/Articles/MGB-Voltage-Stabilizer.htm

My posting to thread Voltage Stabilizer https://www.volvoforums.org.uk/showthread.php?t=284381

"Your symptoms are consistent with a Voltage Stabilizer fault...to locate VStab, have a look at back of Tach or Speedo with a good light...sorry, they hid it pretty well, so some contortions are required...factory typically fastened them as a secondary use for the instrument securing hardware (giving continuous conductive path down to Dashboard metal)...but there might have been variations from the factory or mechanics which came along later...so remember: The Wiring Diagram Color Code is your friend...use wiring colors to verify you have indeed located the VS!...[White-GreeN looks to be input and BLue output for ES]
Be also aware, the Mount (M) hardware also acted as a chassis connection (important to the correct elec function of the VS!) so first thing when you locate it is to loosen, then resnug the securing nut to assure a good chassis connection (poor connection at VS mounting tab and any associated mechanical/electrical conductive path to Dashboard metal will cause malfunction including intermittency, although the "Chassis Tab" (C) of the VS itself simply wraps around the board edge and is held in-place by crimped over VS housing (see VS construction details below):  ...this connection itself is susceptible after such a long time, to corrosion...)...to check function of VS, locate and connect a temporary test wire to the output of VS (best place is at one of the other instruments), and monitor voltage...VS output should be consistent with that shown in link." [Cycling!]

 

-----------------------------------------------

Correspondence with Graham:  

Graham;

Thanks for elaborating and details...the Oil Temp hardly moving is quite typical, because that Temp hardly ever gets into the gauge range anyway...but the Fuel and Temp Gauges high is what I was rather expecting to hear...because I fully understand this mode of failure now...(Fuel Gauge bouncing is likely a Sender or connection to Sender intermittency not associated with V-Stab)...

Good news and Bad news are: I understand and can explain your symptoms, but the old part works fine and didn't actually need replacement. Sorry!

In checking your old V-Stab, I actually found it to work as expected(!), and without a problem. So it remains to explain the symptoms you experienced...I am able to explain what happened: The chassis connection at the mount was indeed less than adequate as I originally suspected...this caused the internal Heating Element to stop working, which keeps the internal contact closed and from cycling, putting a steady 12V on the out put terminal, resulting in the gauges reading high.

Fortunately, the replacement component is not terribly expensive and replacing it with a new one meant the securing nut was tightened once again, making a good chassis connection for the new part, and everything again functions as expected, so the work and expense was not for naught, but indeed, the root cause was again one of the most common problems with cars:  A Bad Ground!

The part is doing its duty now, but I will send you some ACZP to apply to the chassis mounting tab...so if you get bored and adventurous (or in 20 years, the symptoms reoccur for your grandson...!), or you feel like crawling back under your dashboard again now, you can apply it to the mounting stud and thumbscrew to assure a trouble-free contact forever.

You can expect to see info of this experience, in addition to other details, added to the SW-EM V-Stab Tech Article shortly...I will post to thread when info is live...
Please fwd your mailing address so that I can send along the ACZP.

Thanks and Regards, Ron

-----------------------------------------------

Links: 

https://www.autoelectricalspares.co.uk/smiths-br130800-voltage-stabiliser-negtive-earth-top-bracket-1383-p.asp

https://www.spiyda.com/instrument-voltage-stabilizer-kit.html

-----------------------------------------------

Reference and Additional Information:  

https://en.wikipedia.org/wiki/Bimetallic_strip

PLACEHOLDER FOR 140 V-Stab Info (associated with VDO supplied gauges) PN 684623 ("5.1V")

-----------------------------------------------

External material is attributed.  The terms Volvo and Smiths are used here for reference only.  I have no affiliation with either company, other than to try to keep their products including subsystems working for me, and to help other enthusiasts try to do the same.  The results and highly opinionated impressions presented here are from my own carefully considered experience, and can be used (or not, and used strictly for your amusement!) or ridiculed, at your discretion and risk!  As with any recipe, your results may vary, and as always, if you can supply corrections, or additional objective information or experience, I will consider it, and consider working it into the next revision of this article...along with likely the odd metaphor and maybe wise-a** comment. 

You are welcome to use the information here in good health, and for your own non-commercial purposes, but if you reprint or otherwise republish this material, you must give credit to the author or link back to the SwEm site as the source.  If you donít, youíre just a lazy, scum sucking plagiarist...the Washington Post wants you! 

 

B A C K ! . . .to Tech Articles Index Page